Software Engineer, ML Performance
$224k - $357k • US, CA, Santa Clara
Posted 19h ago
About the job
NVIDIA is seeking a dedicated Software Engineer, ML Performance to join our team in Santa Clara, CA. In this role, you will engage directly with the latest AI technologies and work on optimizing performance at an outstanding scale. Your expertise will ensure that our groundbreaking solutions operate flawlessly and continue to lead the industry. You will be immersed in an inclusive, supportive environment where everyone is inspired to do their best work. Come join the team and see how you can make a lasting impact on the world.
Responsibilities
- Analyze the performance of distributed training and inference workloads across the entire stack, identifying bottlenecks and proposing solutions.
- Define performance metrics for a fleet of GPUs and break them down into actionable factors for engineers.
- Drive the analysis of performance regressions by isolating causes and creating reusable analysis methods.
- Evaluate automated diagnoses for accuracy, identify misleading results, and build labeled cases for honest analysis.
- Collaborate with build partners on their workloads, carry findings through to improvements, and provide feedback for product roadmaps.
Requirements
- Bachelor’s degree in Computer Science, Computer Engineering, or related field, or equivalent experience.
- 12+ years of relevant software engineering experience, including hands-on optimization of GPU training or inference workloads in production.
- Demonstrated performance work across multiple layers of the stack with clear documentation of baseline, evidence, alternative explanations, and validated outcomes.
- Experience diagnosing performance across many jobs, machines, or tenants, not just tuning a single workload.
- Familiarity with framework, runtime, and compiler interfaces to identify signal measurement points.
- Ability to distinguish symptoms from causes, recognize noisy or incomplete measurements, and understand the effects of measurement itself.
- Strong programming and debugging skills, demonstrated by maintainable analysis code, tests, or repeatable investigation procedures.
- Clear communication of findings, trade-offs, and uncertainty to workload owners.
- Proficiency with Python and CUDA.
Benefits
- Equity
- Benefits